Advancements in supervised deep learning for metal artifact reduction in computed tomography: A systematic review
Cecile E J Kleber1, Ramez Karius1, Lucas E Naessens1
1Department of Clinical Technology, Faculty of Mechanical Engineering, Delft University of Technology, Delft, the Netherlands.
European Journal of Radiology
|September 12, 2024
Summary
Deep learning algorithms significantly reduce metal artifacts in CT scans, improving image quality. Further standardization is needed for clinical evaluation of these advanced metal artifact reduction (MAR) techniques.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Radiology
Background:
- Metal artifacts in CT imaging degrade image quality and diagnostic accuracy.
- Deep learning (DL) based metal artifact reduction (MAR) algorithms are emerging for clinical use.
Conclusions:
- Supervised DL-MAR algorithms show promise in reducing metal artifacts and enhancing CT image quality.
- Standardized evaluation methodologies are crucial for comparing DL-MAR algorithm performance on clinical data.
